In 1983, millions of unsold cartridges of the Atari game ET The Extra Terrestrial were secretly buried in a concrete-covered landfill in Alamogordo, New Mexico. Thanks to extremely rushed development and alleged interference from Universal Pictures, the movie tie-in was one of the worst video games ever made, and a mass grave was the only option for the poor, unwanted cartridges. Unfortunately, like movies adapted from games, games adapted from movies are often bad. Historically, this has not been the result of too much interference from the movie studios that own the licences, but too little. Promising pop-culture properties such as Transformers, Robocop and Harry Potter have often been farmed out to contracted development studios that are then given far too little time to make anything half decent. The forthcoming Star Wars Jedi: Fallen Order, however, has better prospects than most.

When you're watching a modern blockbuster such as The Avengers, it's hard to escape the feeling that what you're seeing is almost entirely computer-generated imagery, from the effects to the sets to fantastical creatures. But if there's one thing you can rely on to be 100% real, it's the actors. We might have virtual pop stars like Hatsune Miku, but there has never been a world-famous virtual film star. Even that link with corporeal reality, though, is no longer absolute. You may have already seen examples of what's possible: Peter Cushing (or his image) appearing in Rogue One: A Star Wars Story more than 20 years after his death, or Tupac Shakur performing from beyond the grave at Coachella in 2012.

The use of computational methods to evaluate aesthetics in photography has gained interest in recent years due to the popularization of convolutional neural networks and the availability of new annotated datasets. Most studies in this area have focused on designing models that do not take into account individual preferences for the prediction of the aesthetic value of pictures. We propose a model based on residual learning that is capable of learning subjective, user specific preferences over aesthetics in photography, while surpassing the state-of-the-art methods and keeping a limited number of user-specific parameters in the model. Our model can also be used for picture enhancement, and it is suitable for content-based or hybrid recommender systems in which the amount of computational resources is limited.

Emotional aspects play an important part in our interaction with music. However, modelling these aspects in MIR systems have been notoriously challenging since emotion is an inherently abstract and subjective experience, thus making it difficult to quantify or predict in the first place, and to make sense of the predictions in the next. In an attempt to create a model that can give a musically meaningful and intuitive explanation for its predictions, we propose a VGG-style deep neural network that learns to predict emotional characteristics of a musical piece together with (and based on) human-interpretable, mid-level perceptual features. We compare this to predicting emotion directly with an identical network that does not take into account the mid-level features and observe that the loss in predictive performance of going through the mid-level features is surprisingly low, on average. The design of our network allows us to visualize the effects of perceptual features on individual emotion predictions, and we argue that the small loss in performance in going through the mid-level features is justified by the gain in explainability of the predictions.

Multiple sequence to sequence models were used to establish an end-to-end multi-turns proactive dialogue generation agent, with the aid of data augmentation techniques and variant encoder-decoder structure designs. A rank-based ensemble approach was developed for boosting performance. Results indicate that our single model, in average, makes an obvious improvement in the terms of F1-score and BLEU over the baseline by 18.67% on the DuConv dataset. In particular, the ensemble methods further significantly outperform the baseline by 35.85%.

We attempt to make two arguments in this essay. First, through a case study of a mobile phone based voice-media service we have been running in rural central India for more than six years, we describe several implementation complexities we had to navigate towards realizing our intended vision of bringing social development through technology. Most of these complexities arose in the interface of our technology with society, and we argue that even other technology providers can create similar processes to manage this socio-technological interface and ensure intended outcomes from their technology use. We then build our second argument about how to ensure that the organizations behind both market driven technologies and those technologies that are adopted by the state, pay due attention towards responsibly managing the socio-technological interface of their innovations. We advocate for the technology engineers and researchers who work within these organizations, to take up the responsibility and ensure that their labour leads to making the world a better place especially for the poor and marginalized. We outline possible governance structures that can give more voice to the technology developers to push their organizations towards ensuring that responsible outcomes emerge from their technology. We note that the examples we use to build our arguments are limited to contemporary information and communication technology (ICT) platforms used directly by end-users to share content with one another, and hence our argument may not generalize to other ICTs in a straightforward manner.

Research on style transfer and domain translation has clearly demonstrated the ability of deep learning-based algorithms to manipulate images in terms of artistic style. More recently, several attempts have been made to extend such approaches to music (both symbolic and audio) in order to enable transforming musical style in a similar manner. In this study, we focus on symbolic music with the goal of altering the 'style' of a piece while keeping its original 'content'. As opposed to the current methods, which are inherently restricted to be unsupervised due to the lack of 'aligned' data (i.e. the same musical piece played in multiple styles), we develop the first fully supervised algorithm for this task. At the core of our approach lies a synthetic data generation scheme which allows us to produce virtually unlimited amounts of aligned data, and hence avoid the above issue. In view of this data generation scheme, we propose an encoder-decoder model for translating symbolic music accompaniments between a number of different styles. Our experiments show that our models, although trained entirely on synthetic data, are capable of producing musically meaningful accompaniments even for real (non-synthetic) MIDI recordings.

The increasing rate of information pollution on the Web requires novel solutions to tackle that. Question Answering (QA) interfaces are simplified and user-friendly interfaces to access information on the Web. However, similar to other AI applications, they are black boxes which do not manifest the details of the learning or reasoning steps for augmenting an answer. The Explainable Question Answering (XQA) system can alleviate the pain of information pollution where it provides transparency to the underlying computational model and exposes an interface enabling the end-user to access and validate provenance, validity, context, circulation, interpretation, and feedbacks of information. This position paper sheds light on the core concepts, expectations, and challenges in favor of the following questions (i) What is an XQA system?, (ii) Why do we need XQA?, (iii) When do we need XQA? (iv) How to represent the explanations? (iv) How to evaluate XQA systems?

Unsupervised near-duplicate detection has many practical applications ranging from social media analysis and web-scale retrieval, to digital image forensics. It entails running a threshold-limited query on a set of descriptors extracted from the images, with the goal of identifying all possible near-duplicates, while limiting the false positives due to visually similar images. Since the rate of false alarms grows with the dataset size, a very high specificity is thus required, up to $1 - 10^{-9}$ for realistic use cases; this important requirement, however, is often overlooked in literature. In recent years, descriptors based on deep convolutional neural networks have matched or surpassed traditional feature extraction methods in content-based image retrieval tasks. To the best of our knowledge, ours is the first attempt to establish the performance range of deep learning-based descriptors for unsupervised near-duplicate detection on a range of datasets, encompassing a broad spectrum of near-duplicate definitions. We leverage both established and new benchmarks, such as the Mir-Flick Near-Duplicate (MFND) dataset, in which a known ground truth is provided for all possible pairs over a general, large scale image collection. To compare the specificity of different descriptors, we reduce the problem of unsupervised detection to that of binary classification of near-duplicate vs. not-near-duplicate images. The latter can be conveniently characterized using Receiver Operating Curve (ROC). Our findings in general favor the choice of fine-tuning deep convolutional networks, as opposed to using off-the-shelf features, but differences at high specificity settings depend on the dataset and are often small. The best performance was observed on the MFND benchmark, achieving 96\% sensitivity at a false positive rate of $1.43 \times 10^{-6}$.

Automatic cover detection -- the task of finding in an audio database all the covers of one or several query tracks -- has long been seen as a challenging theoretical problem in the MIR community and as an acute practical problem for authors and composers societies. Original algorithms proposed for this task have proven their accuracy on small datasets, but are unable to scale up to modern real-life audio corpora. On the other hand, faster approaches designed to process thousands of pairwise comparisons resulted in lower accuracy, making them unsuitable for practical use. In this work, we propose a neural network architecture that is trained to represent each track as a single embedding vector. The computation burden is therefore left to the embedding extraction -- that can be conducted offline and stored, while the pairwise comparison task reduces to a simple Euclidean distance computation. We further propose to extract each track's embedding out of its dominant melody representation, obtained by another neural network trained for this task. We then show that this architecture improves state-of-the-art accuracy both on small and large datasets, and is able to scale to query databases of thousands of tracks in a few seconds.